BACKGROUND: Intrahepatic arterial yttrium 90 ( 90 Y) microspheres have been proposed as a less toxic, less invasive therapeutic option to transhepatic arterial chemoembolization (TACE) for patients with surgically unresectable hepatocellular carcinoma (HCC). TACE has demonstrated the ability to prolong survival. However, long-term survival remains uncertain. METHODS: In a 2-cohort experience in the treatment of North American patients who had advanced, unresectable, biopsy-proven HCC, 691 patients received repetitive, cisplatin-based chemoembolization; and a separate cohort of 99 patients who had similar treatment criteria received a planned, single dose of 90 Y. Over the study period, an additional 142 patients were followed without treatment (total, 932 patients). RESULTS: Overall survival was slightly better in the 90 Y group compared with the TACE group (median survival, 11.5 months vs 8.5 months). However, the selection criteria indicated a small but significant bias toward milder disease in the 90 Y group. By using stratification into a 3-tier model with patients dichotomized according to bilirubin levels <1.5 mg/dL, the absence of portal vein thrombosis (PVT), and low a-fetoprotein plasma levels (<25 U/dL), an analysis of survival in clinical subgroups indicated that the 2 treatments resulted in similar survival. In addition, patients who had PVT or high afetoprotein levels also had similar survival in both treatment groups. CONCLUSIONS: Given the current evidence of therapeutic equivalence in survival, 90 Y and TACE appeared to be equivalent regional therapies for patients with unresectable, nonmetastatic HCC.